QIIME 2

#!/usr/bin/env python
# -----------------------------------------------------------------------------
# Auto-generated by qiime2 v.2024.5.0.dev0 at 05:54:52 PM on 05 May, 2024
# This document is a representation of the scholarly work of the creator of the
# QIIME 2 Results provided as input to this software, and may be protected by
# intellectual property law. Please respect all copyright restrictions and
# licenses governing the use, modification, and redistribution of this work.

# For User Support, post to the QIIME2 Forum at https://forum.qiime2.org.

# Instructions for use:
# 1. Open this script in a text editor or IDE. Support for Python
#    syntax highlighting is helpful.
# 2. Search or scan visually for '<' or '>' characters to find places where
#    user input (e.g. a filepath or column name) is required. If syntax
#    highlighting is enabled, '<' and '>' will appear as syntax errors.
# 3. Search for 'FIXME' comments in the script, and respond as directed.
# 4. Remove all 'FIXME' comments from the script completely. Failure to do so
#    may result in 'Missing Option' errors
# 5. Adjust the arguments to the commands below to suit your data and metadata.
#    If your data is not identical to that in the replayed analysis,
#    changes may be required. (e.g. sample ids or rarefaction depth)
# 6. Optional: search for 'SAVE' comments in the script, commenting out the
#    `some_result.save` lines for any Results you do not want saved to disk.
# 7. Activate your replay conda environment, and confirm you have installed all
#    plugins used by the script.
# 8. Run this script with `python <path to this script>`, or paste commands
#    into a python interpreter or jupyter notebook for an interactive analysis
# -----------------------------------------------------------------------------

from qiime2 import Artifact
from qiime2 import Metadata
import qiime2.plugins.composition.actions as composition_actions
import qiime2.plugins.dada2.actions as dada2_actions
import qiime2.plugins.demux.actions as demux_actions
import qiime2.plugins.feature_classifier.actions as feature_classifier_actions
import qiime2.plugins.feature_table.actions as feature_table_actions
import qiime2.plugins.rescript.actions as rescript_actions
import qiime2.plugins.taxa.actions as taxa_actions

feature_data_taxonomy_0 = Artifact.import_data(
    'FeatureData[Taxonomy]',
    <your data here>,
)
# SAVE: comment out the following with '# ' to skip saving this Result to disk
feature_data_taxonomy_0.save('feature_data_taxonomy_0')

feature_data_sequence_0 = Artifact.import_data(
    'FeatureData[Sequence]',
    <your data here>,
)
# SAVE: comment out the following with '# ' to skip saving this Result to disk
feature_data_sequence_0.save('feature_data_sequence_0')

emp_single_end_sequences_0 = Artifact.import_data(
    'EMPSingleEndSequences',
    <your data here>,
)
# SAVE: comment out the following with '# ' to skip saving this Result to disk
emp_single_end_sequences_0.save('emp_single_end_sequences_0')

reads_0, = feature_classifier_actions.extract_reads(
    sequences=feature_data_sequence_0,
    f_primer='GTGCCAGCMGCCGCGGTAA',
    r_primer='GGACTACHVGGGTWTCTAAT',
    trim_right=0,
    trunc_len=0,
    trim_left=0,
    identity=0.8,
    min_length=50,
    max_length=0,
    n_jobs=1,
    batch_size='auto',
    read_orientation='both',
)
# SAVE: comment out the following with '# ' to skip saving Results to disk
reads_0.save('reads_0')

# Replay attempts to represent metadata inputs accurately, but metadata .tsv
# files are merged automatically by some interfaces, rendering distinctions
# between file inputs invisible in provenance. We output the recorded metadata
# to disk to enable visual inspection.

# The following command may have received additional metadata .tsv files. To
# confirm you have covered your metadata needs adequately, review the original
# metadata, saved at './recorded_metadata/demux_emp_single_0/'

# NOTE: You may substitute already-loaded Metadata for the following, or cast a
# pandas.DataFrame to Metadata as needed.

barcodes_0_md = Metadata.load(<your metadata filepath>)
barcodes_0_mdc_0_mdc = barcodes_0_md.get_column('<column name>')
per_sample_sequences_0, _ = demux_actions.emp_single(
    seqs=emp_single_end_sequences_0,
    barcodes=barcodes_0_mdc_0_mdc,
    golay_error_correction=True,
    rev_comp_barcodes=False,
    rev_comp_mapping_barcodes=False,
    ignore_description_mismatch=False,
)
# SAVE: comment out the following with '# ' to skip saving Results to disk
per_sample_sequences_0.save('per_sample_sequences_0')

dereplicated_sequences_0, dereplicated_taxa_0 = rescript_actions.dereplicate(
    sequences=reads_0,
    taxa=feature_data_taxonomy_0,
    mode='uniq',
    perc_identity=1.0,
    threads=1,
    rank_handles='greengenes',
    derep_prefix=False,
)
# SAVE: comment out the following with '# ' to skip saving Results to disk
dereplicated_taxa_0.save('dereplicated_taxa_0')
dereplicated_sequences_0.save('dereplicated_sequences_0')

table_0, representative_sequences_0, _ = dada2_actions.denoise_single(
    demultiplexed_seqs=per_sample_sequences_0,
    trunc_len=120,
    trim_left=0,
    max_ee=2.0,
    trunc_q=2,
    pooling_method='independent',
    chimera_method='consensus',
    min_fold_parent_over_abundance=1.0,
    n_threads=1,
    n_reads_learn=1000000,
    hashed_feature_ids=True,
)
# SAVE: comment out the following with '# ' to skip saving Results to disk
table_0.save('table_0')
representative_sequences_0.save('representative_sequences_0')

classifier_0, _, _ = rescript_actions.evaluate_fit_classifier(
    sequences=dereplicated_sequences_0,
    taxonomy=dereplicated_taxa_0,
    reads_per_batch='auto',
    n_jobs=1,
    confidence=0.7,
)
# SAVE: comment out the following with '# ' to skip saving Results to disk
classifier_0.save('classifier_0')

# The following command may have received additional metadata .tsv files. To
# confirm you have covered your metadata needs adequately, review the original
# metadata, saved at './recorded_metadata/feature_table_filter_samples_0/'

# NOTE: You may substitute already-loaded Metadata for the following, or cast a
# pandas.DataFrame to Metadata as needed.

metadata_0_md = Metadata.load(<your metadata filepath>)
filtered_table_0, = feature_table_actions.filter_samples(
    table=table_0,
    min_frequency=0,
    min_features=0,
    metadata=metadata_0_md,
    where="[body-site]='gut'",
    exclude_ids=False,
    filter_empty_features=True,
)
# SAVE: comment out the following with '# ' to skip saving Results to disk
filtered_table_0.save('filtered_table_0')

classification_0, = feature_classifier_actions.classify_sklearn(
    reads=representative_sequences_0,
    classifier=classifier_0,
    reads_per_batch='auto',
    n_jobs=1,
    pre_dispatch='2*n_jobs',
    confidence=0.7,
    read_orientation='auto',
)
# SAVE: comment out the following with '# ' to skip saving Results to disk
classification_0.save('classification_0')

collapsed_table_0, = taxa_actions.collapse(
    table=filtered_table_0,
    taxonomy=classification_0,
    level=6,
)
# SAVE: comment out the following with '# ' to skip saving Results to disk
collapsed_table_0.save('collapsed_table_0')

composition_table_0, = composition_actions.add_pseudocount(
    table=collapsed_table_0,
    pseudocount=1,
)
# SAVE: comment out the following with '# ' to skip saving Results to disk
composition_table_0.save('composition_table_0')

# The following command may have received additional metadata .tsv files. To
# confirm you have covered your metadata needs adequately, review the original
# metadata, saved at './recorded_metadata/composition_ancom_0/'

# NOTE: You may substitute already-loaded Metadata for the following, or cast a
# pandas.DataFrame to Metadata as needed.

metadata_1_md = Metadata.load(<your metadata filepath>)
metadata_1_mdc_0_mdc = metadata_1_md.get_column('<column name>')
visualization_0_viz, = composition_actions.ancom(
    table=composition_table_0,
    metadata=metadata_1_mdc_0_mdc,
    transform_function='clr',
)
# SAVE: comment out the following with '# ' to skip saving Results to disk
visualization_0_viz.save('visualization_0_viz')


# -----------------------------------------------------------------------------
# The following QIIME 2 Results were parsed to produce this script:
# e347166b-8741-46bc-b6de-f1021b66d3b6
# -----------------------------------------------------------------------------